ToxSci Advance Access originally published online on February 19, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Toxicological Sciences 79, 38-40 (2004)
Toxicological Sciences vol. 79 no. 1 © Society of Toxicology; all rights reserved.
Correlation of Tumors with DNA Adducts from Methyl Eugenol and Tamoxifen in Rats
* Department of Pharmacology and Toxicology, School of Medicine, University of Louisville, Louisville, Kentucky; and Biological Chemistry, Division of Biomedical Sciences, Faculty of Medicine, Imperial College London, London, SW72AZ, UK
Received December 29, 2003; accepted January 23, 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Data on percent tumors in male rats after administration of methyl eugenol, obtained from the National Toxicology Program, or tamoxifen were plotted on a linear scale for percent tumors against the dose on a logarithmic scale. Data on 32P-postlabelled DNA adducts were plotted on the same graphs for each of these two compounds in order to correlate adduct formation and tumor incidence with dose. The resulting graph for methyl eugenol showed a linear response for both adduct formation and tumor incidence. The threshold dose of administered methyl eugenol for adduct formation (zero adducts) was 1019.3 molecules of methyl eugenol/kg/day, which compared with a threshold of 1020.1 molecules of methyl eugenol/kg/day for tumor formation; however, 30 adducts/108 nucleotides was the threshold for tumor formation. The dose of tamoxifen for adduct formation fit an exponential plot slightly better than a linear plot, but reached minimal values close to the threshold of 1018.7 molecules of tamoxifen/kg/day for tumor formation. These data confirm that tumor formation coincides with adduct formation and that both have thresholds, or at least reach minimal values, above levels to which humans are exposed. Although the threshold dose for tumor formation from tamoxifen is only about 10x above the dose received by women at risk for breast cancer, this should be an adequate safety margin. The safety factor for methyl eugenol is several orders of magnitude; therefore, there should be no cause for concern for humans at current levels of exposure.
Key Words: methyl eugenol; tamoxifen; carcinogenesis; thresholds; DNA adducts; 32P-postlabelling.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methyl eugenol is a naturally occurring flavor in basil, allspice, nutmeg, and other edible plants. It also has been found to be carcinogenic in F344 male rats when given at high doses for a lifetime (NTP TR 491). A previous publication (Waddell, 2002
) demonstrated a clear threshold for the carcinogenicity of methyl eugenol in rats that is orders of magnitude above the levels consumed by humans. Other data on DNA adduct formation in the site of tumors (liver) in rats have become available (Carmichael et al., 1999); however, a dose-response evaluation for adduct formation has not been published.
Tamoxifen is a synthetic "antiestrogen" that is in current clinical use in women for the treatment and prevention of breast cancer. Tamoxifen also has been shown to be carcinogenic in rats when given at high doses (Greaves et al., 1993). Data also are available on DNA adduct formation in the site of tumors (liver) in rats (White et al., 1992); however, a dose-response evaluation for adduct formation and tumor incidence in rats has been not been published.
This report evaluates the dose response for DNA adduct formation from methyl eugenol and compares that threshold with the carcinogenicity threshold and also with doses to which humans are exposed. The DNA adduct formation for tamoxifen is also compared with the carcinogenicity threshold for tamoxifen in rats and the dose of tamoxifen recommended for women who are at high risk for breast cancer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Threshold calculations and human exposure data on methyl eugenol were taken from a previous publication (Waddell, 2002
). DNA adduct data on methyl eugenol are 32P-postlabelling data (Carmichael et al., 1999); these were plotted on the same graph prepared using SlideWrite software (Advanced Graphics Software, Inc., Encinitas, CA). The regression statistics were calculated by the SlideWrite software. Justifications for calculating thresholds based on a log-linear plot using the Rozman scale (Rozman et al., 1996) are provided in that (Waddell, 2002) and subsequent publications (Rozman, 2003; Waddell, 2003a,b,c,d; Waddell and Bates, 1969).
Dose-response calculations for tumor formation and DNA adduct formation from tamoxifen were prepared by the same SlideWrite software and on the same scale from data published by Greaves et al. (1993) and White et al. (1992). The dose recommended for women at high risk for breast cancer of 20 mg per day (PDR, 2003) was converted to molecules/kg/day assuming a woman of 60 kg body weight and plotted on the same graph with the tamoxifen thresholds.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Figure 1 shows DNA postlabelling adducts on the same graph previously published that presents tumor threshold data along with doses consumed by different categories of eaters. Details of consumption by categories of eaters can be found in the previous publication (Waddell, 2002
). It appears that adduct formation begins at a dose to rats of about 1019.3 molecules of methyl eugenol/kg/day. This is about one order of magnitude below the dose that produces tumors in rats. DNA adduct data had a slightly better fit when plotted on a linear scale (r = 0.993) than when plotted on an exponential scale (r = 0.990). Regardless of the shape of the dose-response curve for adduct formation, superimposing grids from SlideWrite software reveal that tumor formation did not begin until there were about 30 postlabelling adducts/108 nucleotides.
|
Figure 2 shows the data for tumor and adduct formation from tamoxifen along with the recommended dosage for women at high risk for breast cancer. DNA adduct data had a slightly better fit when plotted on an exponential scale (r = 0.988) than when plotted on a linear scale (r = 0.985). Both the threshold with a linear plot and minimal values with the exponential plot coincide closely with the threshold for tumor formation; this corresponds to a dose of about 1018.7 molecules of tamoxifen/day. This threshold for tumor formation is about 10x higher than the dose recommended for high-risk women. The exponential curve shown in Figure 2 is the calculated best fit for the adduct data. It has a value of 5.73 postlabelling adducts/108 nucleotides (as calculated by the SlideWrite software) at the threshold for tumor formation (i.e., 1018.7).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Direct comparison of DNA adduct formation with tumor incidence is obviously of great interest and has been attempted in numerous publications. However, to our knowledge, none of these publications has been able to correlate these two events with their thresholds. Previous publications (Rozman, 2003
; Waddell, 2002; Waddell, 2003a,b,c,d; Waddell and Bates, 1969) have shown that a log scale for dose and a linear scale for quantal response is the only scale consistent with the fundamental laws of chemistry and that use of such a scale can readily reveal thresholds. The Rozman scale (Rozman et al., 1996) is ideal for this plot because, in addition to being log linear, it is continuous to one molecule and uses molecules instead of weight of chemical. Indeed, in this report, when data for DNA adduct formation were compared with tumor formation in rats, using this scale, there is a remarkable correlation between the thresholds for these events. The calculations in this report suffer somewhat from not being done in the same experiment. Although the tumor incidence and adduct data from methyl eugenol were both determined in F-344 male rats, the tamoxifen data were obtained from F-344 female rats (adducts) and Wistar-derived rats (tumors). In addition, the length of treatment for the adduct data differed from the treatment duration for adducts. For example, the tamoxifen tumor data were obtained from a two-year carcinogenicity study (the first tumors being observed after 31 weeks of treatment in the top dose group), whereas the appearance of DNA adducts was measured over a seven-day dosing regime. Furthermore, 32P-postlabelling may or may not be an accurate measure of actual adducts. Comparisons with direct analyses of adducts are planned in future reports.
Superimposing a vertical grid on Figure 1 revealed that tumor formation from methyl eugenol did not begin until there were about 30 adducts/108 nucleotides. This threshold for tumor formation compares with about 30 adducts/106 nucleotides found in B6C3F1 mice by Phillips et al. (1984). However, the data are not firm because they are in different species and from different experiments. The same is true for the tamoxifen data and, although the DNA adducts fit an exponential plot better than a linear plot, minimal values of adducts coincide with the threshold for tumor formation. It appears that a dose to the animals of about 1018.7 molecules/day of tamoxifen is required for tumor formation to begin. At this threshold for tumor formation the exponential fit for adducts predicts that 5.73 adducts/108 nucleotides are present. This would suggest that adducts below this value do not produce tumors. DNA repair is a remarkably efficient process; consequently, the thresholds seen in this report arise when those repair mechanisms are overwhelmed and tumor formation actually is seen.
Tumor formation from both methyl eugenol and tamoxifen appears to be linear with the logarithm of the dose. The data for DNA adduct formation, however, are not sufficient to distinguish between a linear response and an exponential response with the logarithm of the dose. In the case of methyl eugenol, a linear response gives a slightly better fit, but in the case of tamoxifen, an exponential response gives a slightly better fit. Resolution of this question awaits further data. In the event that DNA adducts do actually prove eventually to increase exponentially with the logarithm of the dose, our suggestion at this time is that adducts might follow a Hill-type equation. That is, as adducts form, the DNA molecule becomes more accessible to further adducts.
The publication by Carthew, P., Lee, P. N., Edwards, R. E., Heydon, R. T., Nolan, B. M., and Martin, E. A. (2001). Cumulative exposure to tamoxifen: DNA adducts and liver cancer in the rat. Arch. Toxicol. 75:375–380 has just come to our attention. Those authors found that DNA adducts from tamoxifen below 180 postlabelling adducts per 108 nucleotides did not produce liver tumors in Wistar rats. The difference between that threshold and our calculated threshold of about 6 nucleotides per 108 nucleotides could be due to differences in strains of rats or dosing schedules. We apologize for overlooking this publication with which the present report is consistent and supportive.
|
ACKNOWLEDGMENTS |
|---|
Supported, in part, by Pharmacon Research Foundation, Inc.
|
NOTES |
|---|
1 To whom correspondence should be addressed at 14300 Rose Wycombe Lane, Prospect, KY 40059. Fax: (502) 228-6779. E-mail: bwaddell{at}louisville.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Carmichael, P. L., Crooks, N., Schenk, M., Gardner, I., Kenna, J. G., and Caldwell, J. (1999). A comparison of immunochemical and postlabelling methodologies for the detection of DNA adducts formed by methyleugenol. Hum. Exp. Toxicol. 18, 55.
Greaves, P., Goonetilleke, R., Nunn, G., Topham, J., and Orton, T. (1993). Two-year carcinogenicity study of tamoxifen in Alderley Park Wistar-derived rats. Cancer Res. 53, 3919–3924.
NTP (2000). Toxicology and carcinogenises studies of methyleugenol in F344/N rats and B6C3F; mice. National Toxicology Program, NTP TR 491. Available online at http://ntp-server.niehs.nih.gov.
PDR (2003). Physicians’ Desk Reference, 57th ed. Thomson PDR, Montvale, NJ.
Phillips, D. H., Reddy, M. V., and Randerath, K. (1984). 32P-Post-labelling analysis of DNA adducts formed in the livers of animals treated with safrole, estragole and other naturally occurring alkenylbenzenes. II. Newborn male B6C3F1 mice. Carcinogenesis 5, 1623–1628.
Rozman, K. K., Kerecsen, L., Viluksela, M. K., Österle, D., Deml, E., Viluksela, M., Stahl, B. U., Greim, H., and Doull, J. (1996). A toxicologist’s view of cancer risk assessment. Drug Metab. Rev. 28, 29–52.[Web of Science][Medline]
Rozman, K. K. (2003). Letter to the editor. Toxicol. Sci. 75, 224–225.
Waddell, W. J. (2002). Thresholds of carcinogenicity of flavors. Toxicol. Sci. 68, 275–279.
Waddell, W. J. (2003a). Thresholds of carcinogenicity in the ED01 study. Toxicol. Sci. 72, 158–163.
Waddell, W. J. (2003b). Threshold of carcinogenicity of nitrosodiethylamine for esophageal tumors in rats. Food Chem. Toxicol. 41,739–741.[CrossRef][Web of Science][Medline]
Waddell, W. J. (2003c). Thresholds in chemical carcinogenesis: What are animal experiments telling us? Toxicol. Pathol. 31, 260–262.[CrossRef][Web of Science][Medline]
Waddell, W. J. (2003d). Comparison of human exposures to selected chemicals with thresholds from NTP carcinogenicity studies in rodents. Hum. Exp. Toxicol. 22, 501–506.
Waddell, W. J., and Bates, R. G. (1969). Intracellular pH. Physiol. Rev. 49, 285–329.
White, I. N. H., de Matteis, F., Davies, A., Smith, L. L., Crofton-Sleigh, C., Venitt, S., Hewer, A., and Phillips, D. H. (1992). Genotoxic potential of tamoxifen and analogues in female Fischer F344/n rats, DBA/2 and C57BL/6 mice and in human MCL-5 cells. Carcinogenesis 13, 2197–2203.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  W J Waddell Critique of dose response in carcinogenesis Human and Experimental Toxicology, July 1, 2006; 25(7): 413 - 436. [Abstract] [PDF]
|
|
|
|
|
|
W J Waddell Waddell reply to letter of Andersen and Renwick Human and Experimental Toxicology, November 1, 2005; 24(11): 605 - 606. [PDF]
|
|
|
|
|
|
W. J Waddell Comparisons of thresholds for carcinogenicity on linear and logarithmic dosage scales Human and Experimental Toxicology, June 1, 2005; 24(6): 325 - 332. [Abstract] [PDF]
|
|
|
|
 |
|
 F. L. Martin and K. T. Semple Environmental health impacts: occurrence, exposure and significance, Lancaster University, UK, 9-10 September 2003 Mutagenesis, September 1, 2004; 19(5): 423 - 429. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||